Methods and Systems for Constructing Map Data Using Poisson Surface Reconstruction

1. A method for generating a mesh representation of a surface comprising: receiving a three-dimensional (3D) point cloud representing the surface; identifying and discarding one or more outliers in the 3D point cloud to generate a filtered point cloud using a Gaussian process; adding one or more additional points to the filtered point cloud to generate a reconstruction dataset; using Poisson surface reconstruction to generate an implicit surface corresponding to the surface from the reconstruction dataset; and generating, using the implicit surface, a polygon mesh representation of the surface by using a fine-to-coarse hash map for building polygons at a highest resolution first followed by progressively coarser resolution polygons.

2. The method of claim 1 , further comprising using the polygon mesh representation for navigating an autonomous vehicle over the surface.

3. The method of claim 1 , wherein the polygon mesh representation is a triangle mesh representation.

4. The method of claim 1 , wherein generating the polygon mesh representation comprises using at least one of the following: a marching cube algorithm, a marching tetrahedrons algorithm, or a Bloomenthal Polygonizer.

5. The method of claim 1 , wherein generating the polygon mesh representation comprises detecting and filling holes in the polygon mesh representation by: generating a list of edges that comprises a plurality of edges which have been used once in the polygon mesh; and building polygons starting from a first vertex in the implicit surface using the list of edges until reaching the first vertex such that each of the plurality of edges is used twice in the polygon mesh.

6. The method of claim 1 , wherein identifying and discarding the one or more outliers in the 3D point cloud to generate the filtered point cloud using the Gaussian process comprises: identifying a Gaussian surface corresponding to the 3D point cloud; determining a mean Gaussian surface from the Gaussian surface; and identifying the one or more outliers as points in the 3D point cloud that have a standard deviation from the mean Gaussian surface that is greater than a threshold standard deviation.

7. The method of claim 1 , wherein identifying and discarding the one or more outliers in the 3D point cloud to generate the filtered point cloud using the Gaussian process comprises: identifying a Gaussian surface corresponding to the 3D point cloud; determining a mean Gaussian surface from the Gaussian surface; and identifying the one or more outliers as points in the 3D point cloud that are located at a physical distance from the mean Gaussian surface that is greater than a threshold physical distance.

8. The method of claim 1 , wherein adding the one or more additional points to the filtered point cloud to generate the reconstruction dataset comprises: identifying one or more holes in the filtered point cloud; and adding at least one point to each of the one or more holes using a Gaussian surface corresponding to the 3D point cloud.

9. The method of claim 8 , wherein identifying the one or more holes comprises: grid-sampling a subset of points in the filtered point cloud to determine whether at least one point of the point cloud exists within a threshold distance from a sampled point; and identifying a hole proximate to the sampled point upon determining that at least one point of the point cloud does not exist within the threshold distance from the sampled point.

10. The method of claim 9 , further comprising adding the sampled point to the reconstruction dataset upon determining that least one point of the point cloud exists within the threshold distance from the sampled point.

11. The method of claim 1 , wherein receiving the 3D point cloud representing the surface comprises generating the 3D point cloud by mapping the surface using a mapping sensor to generate sensor data and using the sensor data for generating the 3D point cloud.

12. A system comprising: one or more mapping sensors; a processor; and a non-transitory computer readable medium comprising one or more programming instructions that when executed by the processor, cause the processor to: receive, a three-dimensional (3D) point cloud representing a surface, identify and discard one or more outliers in the 3D point cloud to generate a filtered point cloud using a Gaussian process, add one or more additional points to the filtered point cloud to generate a reconstruction dataset, use Poisson surface reconstruction to generate an implicit surface corresponding to the surface from the reconstruction dataset, and generate, using the implicit surface, a polygon mesh representation of the surface by using a fine-to-coarse hash map for building polygons at a highest resolution first followed by progressively coarser resolution polygons.

13. The system of claim 12 , further comprising programming instructions that when executed by the processor, cause the processor to use the polygon mesh representation for navigating an autonomous vehicle over the surface.

14. The system of claim 12 , wherein the polygon mesh representation is a triangle mesh representation.

15. The system of claim 12 , wherein the instructions that cause the processor to generate the polygon mesh representation comprise instructions to generate the polygon mesh representation by using at least one of the following: a marching cube algorithm, a marching tetrahedrons algorithm, or a Bloomenthal Polygonizer.

16. The system of claim 12 , wherein the instructions that cause the processor to generate the polygon mesh representation further comprise instructions to detect and fill holes in the polygon mesh representation by: generating a list of edges that comprises a plurality of edges which have been used once in the polygon mesh; and building polygons starting from a first vertex in the implicit surface using the list of edges until reaching the first vertex such that each of the plurality of edges is used twice in the polygon mesh.

17. The system of claim 12 , wherein the instructions that cause the processor to identify and discard the one or more outliers in the 3D point cloud to generate the filtered point cloud using the Gaussian process comprise instructions to: identify a Gaussian surface corresponding to the 3D point cloud; determine a mean Gaussian surface from the Gaussian surface; and identify the one or more outliers as points in the 3D point cloud that have a standard deviation from the mean Gaussian surface that is greater than a threshold standard deviation.

18. The system of claim 12 , wherein the instructions that cause the processor to identify and discard the one or more outliers in the 3D point cloud to generate the filtered point cloud using the Gaussian process comprise instructions to: identifying a Gaussian surface corresponding to the 3D point cloud; determining a mean Gaussian surface from the Gaussian surface; and identifying the one or more outliers as points in the 3D point cloud that are located at a physical distance from the mean Gaussian surface that is greater than a threshold physical distance.

19. The system of claim 12 , wherein the instructions that cause the processor to add the one or more additional points to the filtered point cloud to generate the reconstruction dataset comprise instructions to: identify one or more holes in the filtered point cloud; and add at least one point to each of the one or more holes using a Gaussian surface corresponding to the 3D point cloud.

20. The system of claim 19 , wherein the instructions that cause the processor to identify the one or more holes comprise instructions to: grid-sample a subset of points in the filtered point cloud to determine whether at least one point of the point cloud exists within a threshold distance from a sampled point; and identify a hole proximate to the sampled point upon determining that at least one point of the point cloud does not exist within the threshold distance from the sampled point.

21. The system of claim 20 , further comprising programming instructions that when executed by the processor, cause the processor to add the sampled point to the reconstruction dataset upon determining that least one point of the point cloud exists within the threshold distance from the sampled point.

22. The system of claim 12 , wherein the instructions that cause the processor to receive the 3D point cloud representing the surface comprise instructions to generate the 3D point cloud by mapping the surface using a mapping sensor to generate sensor data and using the sensor data for generating the 3D point cloud.

23. A computer program product comprising a non-transitory memory and programming instructions that are configured to cause a processor to: receive, a three-dimensional (3D) point cloud representing a surface, identify and discard one or more outliers in the 3D point cloud to generate a filtered point cloud using a Gaussian process, add one or more additional points to the filtered point cloud to generate a reconstruction dataset, use Poisson surface reconstruction to generate an implicit surface corresponding to the surface from the reconstruction dataset, and generate, using the implicit surface, a polygon mesh representation of the surface by using a fine-to-coarse hash map for building polygons at a highest resolution first followed by progressively coarser resolution polygons.

24. The computer program product of claim 23 , further comprising programming instructions that are configured to cause the processor to use the polygon mesh representation for navigating an autonomous vehicle over the surface.

25. The computer program product of claim 23 , wherein the polygon mesh representation is a triangle mesh representation.

26. The computer program product of claim 23 , wherein the instructions that are configured to cause the processor to generate the polygon mesh representation comprise instructions to generate the polygon mesh representation by using at least one of the following: a marching cube algorithm, a marching tetrahedrons algorithm, or a Bloomenthal Polygonizer.

27. The computer program product of claim 23 , wherein the instructions that are configured to cause the processor to generate the polygon mesh representation further comprise instructions to detect and fill holes in the polygon mesh representation by: generating a list of edges that comprises a plurality of edges which have been used once in the polygon mesh; and building polygons starting from a first vertex in the implicit surface using the list of edges until reaching the first vertex such that each of the plurality of edges is used twice in the polygon mesh.

28. The computer program product of claim 23 , wherein the instructions that are configured to cause the processor to identify and discard the one or more outliers in the 3D point cloud to generate the filtered point cloud using the Gaussian process comprise instructions to: identify a Gaussian surface corresponding to the 3D point cloud; determine a mean Gaussian surface from the Gaussian surface; and identify the one or more outliers as points in the 3D point cloud that have a standard deviation from the mean Gaussian surface that is greater than a threshold standard deviation.

29. The computer program product of claim 23 , wherein the instructions that are configured to cause the processor to identify and discard the one or more outliers in the 3D point cloud to generate the filtered point cloud using the Gaussian process comprise instructions to: identifying a Gaussian surface corresponding to the 3D point cloud; determining a mean Gaussian surface from the Gaussian surface; and identifying the one or more outliers as points in the 3D point cloud that are located at a physical distance from the mean Gaussian surface that is greater than a threshold physical distance.

30. The computer program product of claim 23 , wherein the instructions that are configured to cause the processor to add the one or more additional points to the filtered point cloud to generate the reconstruction dataset comprise instructions to: identify one or more holes in the filtered point cloud; and add at least one point to each of the one or more holes using a Gaussian surface corresponding to the 3D point cloud.

31. The computer program product of claim 30 , wherein the instructions that are configured to cause the processor to identify the one or more holes comprise instructions to: grid-sample a subset of points in the filtered point cloud to determine whether at least one point of the point cloud exists within a threshold distance from a sampled point; and identify a hole proximate to the sampled point upon determining that at least one point of the point cloud does not exist within the threshold distance from the sampled point.

32. The computer program product of claim 31 , further comprising programming instructions that are configured to cause the processor to add the sampled point to the reconstruction dataset upon determining that least one point of the point cloud exists within the threshold distance from the sampled point.

33. The computer program product of claim 23 , wherein the instructions that are configured to cause the processor to receive the 3D point cloud representing the surface comprise instructions to generate the 3D point cloud by mapping the surface using a mapping sensor to generate sensor data and using the sensor data for generating the 3D point cloud.